Prosodic Features and Formant Contribution for Arabic Speech Recognition in Noisy Environments

This paper investigates the contribution of formants and prosodic features like pitch and energy in Arabic speech recognition under real-life conditions. Our speech recognition system based on Hidden Markov Model (HMM) is implemented using the HTK Toolkit. The front-end of the system combines features based on conventional Mel-Frequency Cepstral Coefficient (MFFC), prosodic information and formants. The obtained results show that the resulting multivariate feature vectors lead to a significant improvement of the recognition system performance in noisy environment compared to cepstral system alone.